Targeting disease-causing gene sequences was first suggested more than thirty years ago (Belikova et al., Tet. Lett., 1967, 37, 3557-3562), and antisense activity was demonstrated in cell culture more than a decade later (Zamecnik et al., Proc. Natl. Acad. Sci. U.S.A., 1978, 75, 280-284). One advantage of antisense technology in the treatment of a disease or condition that stems from a disease-causing gene is that it is a direct genetic approach that has the ability to modulate (increase or decrease) the expression of specific disease-causing genes. Another advantage is that validation of a therapeutic target using antisense compounds results in direct and immediate discovery of the drug candidate; the antisense compound is the potential therapeutic agent.
Generally, the principle behind antisense technology is that an antisense compound hybridizes to a target nucleic acid and modulates gene expression activities or function, such as transcription or translation. The modulation of gene expression can be achieved by, for example, target degradation or occupancy-based inhibition. An example of modulation of RNA target function by degradation is RNase H-based degradation of the target RNA upon hybridization with a DNA-like antisense compound. Another example of modulation of gene expression by target degradation is RNA interference (RNAi). RNAi generally refers to antisense-mediated gene silencing involving the introduction of dsRNA leading to the sequence-specific reduction of targeted endogenous mRNA levels. Regardless of the specific mechanism, this sequence-specificity makes antisense compounds extremely attractive as tools for target validation and gene functionalization, as well as therapeutics to selectively modulate the expression of genes involved in the pathogenesis of malignancies and other diseases.
Antisense compounds have been employed as therapeutic agents in the treatment of disease states in animals, including humans. Antisense oligonucleotide drugs are being safely and effectively administered to humans in numerous clinical trials. In 1998, the antisense compound, Vitravene® (fomivirsen; developed by Isis Pharmaceuticals Inc., Carlsbad, Calif.) was the first antisense drug to achieve marketing clearance from the U.S. Food and Drug Administration (FDA), and is currently used in the treatment of cytomegalovirus (CMV)-induced retinitis in AIDS patients. A New Drug Application (NDA) for Genasense™ (oblimersen sodium; developed by Genta, Inc., Berkeley Heights, N.J.), an antisense compound which targets the Bcl-2 mRNA overexpressed in many cancers, was accepted by the FDA. Many other antisense compounds are in clinical trials, including those targeting c-myc (NeuGene® AVI-4126, AVI BioPharma, Ridgefield Park, N.J.), TNF-alpha (ISIS 104838, developed by Isis Pharmaceuticals, Inc.), VLA4 (ATL1102, Antisense Therapeutics Ltd., Toorak, Victoria, Australia) and DNA methyltransferase (MG98, developed by MGI Pharma, Bloomington, Minn.), to name a few.
New chemical modifications have improved the potency and efficacy of antisense compounds, uncovering the potential for oral delivery as well as enhancing subcutaneous administration, decreasing potential for side effects, and leading to improvements in patient convenience. Chemical modifications increasing potency of antisense compounds allow administration of lower doses, which reduces the potential for toxicity, as well as decreasing overall cost of therapy. Modifications increasing the resistance to degradation result in slower clearance from the body, allowing for less frequent dosing. Different types of chemical modifications can be combined in one compound to further optimize the compound's efficacy.
Much of the information regarding the biogenesis of the small nuclear ribonucleoproteins (snRNPs) came from studies of spinal muscular atrophy (SMA). SMA, a motor neuron degenerative disease that results from deletions or mutations in the Survival of Motor Neurons (SMN) gene, is an autosomal recessive disease that is the leading hereditary cause of infant mortality (Gubitz et al., Exp. Cell. Res., 2004, 296, 51-56). The SMN protein is present in the cytoplasm and nucleus, where it is enriched within discrete bodies called Gems (for “Gemini of Cajal bodies”) which are related to and often associated with Cajal bodies (Gubitz et al., Exp. Cell. Res., 2004, 296, 51-56; Liu and Dreyfuss, Embo J., 1996, 15, 3555-3565; Yong et al., Trends Cell Biol., 2004, 14, 226-232). Cajal bodies are known to contain high levels of factors involved in the transcription and processing of many types of nuclear RNAs, including snRNPs, nucleolar ribonucleoproteins (snoRNPs), and the three eukaryotic RNA polymerases, and are most likely sites of assembly and modification of the nuclear transcription and RNA machinery (Gubitz et al., Exp. Cell. Res., 2004, 296, 51-56).
The snRNP particles are components of the spliceosome, the eukaryotic pre-mRNA splicing machinery. Each major snRNP contains a small nuclear RNA (snRNA) as well as a common set of Sm proteins and a set of proteins specific to the particular snRNA. The common Sm proteins are arranged into a core on a uridine-rich sequence in the cytoplasm after nuclear export of the nascent snRNAs. Proper assembly of the core is required for subsequent import of the snRNPs into the nucleus. As compared to other RNP complexes, such as the small nucleolar RNPs which are assembled in the nucleus where they function, the assembly of snRNPs appears to be strictly regulated and complex (Yong et al., Trends Cell Biol., 2004, 14, 226-232).
The SMN protein oligomerizes with a group of proteins named the Gemins. The Gemins include Gemin2, Gemin3, a DEAD/H box helicase, Gemin4, Gemin5, Gemin6, and Gemin7. The Gemins colocalize with SMN in gems and are present in the cytoplasm and nucleoplasm (Gubitz et al., Exp. Cell. Res., 2004, 296, 51-56). It appears that individual Gemins of the SMN complex interact with distinct sets of Sm proteins indicating that multiple contacts are likely to be important for the function of the SMN complex in snRNP core assembly (Baccon et al., J. Biol. Chem., 2002, 277, 31957-31962). The SMN complex was found to function as a specificity factor for the assembly of spliceosomal snRNP, ensuring that Sm cores are only formed on the correct RNA molecules (Pellizzoni et al., Science, 2002, 298, 1775-1779).
Gemin2 (also known as SIP1, SMP-interacting protein 1, and survival of motor neuron protein interacting protein 1) was isolated in a yeast two-hybrid screen of a HeLa cell library using SMN as bait. These proteins were tightly associated and colocalized to Gems, prompting a search for other protein components of the complex. The SMN/Gemin2 pair was found to interact directly with several of the snRNP Sm core proteins (Liu et al., Cell, 1997, 90, 1013-1021). Further implicating the pair in snRNP biogenesis, the SMN/Gemin2 complex associated with spliceosomal snRNAs U1 and U5 in the cytoplasm of Xenopus oocytes, and antibodies against Gemin2 inhibited Sm core assembly of the spliceosomal snRNPs U1, U2, U4, and U5 and their transport to the nucleus (Fischer et al., Cell, 1997, 90, 1023-1029).
Together with Gemin2, Gemin3 and Gemin4 (also known as GIP1 or gem-associated protein 4) were also found to complex with SMN and snRNP proteins (Charroux et al., J. Cell Biol., 1999, 147, 1181-1194; Charroux et al., J. Cell Biol., 2000, 148, 1177-1186). Gemin3 was the only protein of the SMN complex that bound specifically to GST-Gemin4, suggesting that the presence of Gemin4 in the complex is a result of its direct interaction with Gemin3, but not with SMN. The direct and avid interaction of Gemin4 with the DEAD/H box-containing helicase protein Gemin3 may indicate that they function together. Gemin4 also interacts with several of the core Sm proteins, and co-localizes with SMN to gems. Gemin4 localizes to the nucleolus, potentially indicating additional functions in ribosome biogenesis (Charroux et al., J. Cell Biol., 2000, 148, 1177-1186).
Gemin4 proteins are found predominantly in the SMN complex, however, a less abundant Gemin3-Gemin4 complex has also been found (Charroux et al., J. Cell Biol., 2000, 148, 1177-1186; Mourelatos et al., Genes & Development, 2002, 16, 720-728). Immunoprecipitation studies showed that Gemin3 and Gemin4 are associated in a complex with eIF2c2, a member of the large Argonaute family of proteins, members of which have been implicated in RNA interference (RNAi) mechanisms and developmental regulation by short temporal RNAs (stRNAs). The complex, a miRNP, also contained RNAs about 22 nucleotides in length, corresponding to microRNAs (miRNAs). 40 miRNAs were captured in these studies (Mourelatos et al., Genes & Development, 2002, 16, 720-728). Monoclonal antibodies to either Gemin3 or Gemin4 immunoprecipitated let-7-programmed RNA-induced silencing complex (RISC) activity, leading to the notion that the Gemin-4-containing miRNP may be the human RISC which can carry out both target cleavage in the RNAi pathway and translational control in the miRNA pathway (Hutvagner and Zamore, Science, 2002, 297, 2056-2060).
In a yeast-two hybrid assay, Gemin4 was found to interact with galectin-1 and galectin-3, nuclear-localized proteins that were shown to be required factors for splicing in a cell-free assay. This interaction is thought to be functionally relevant in the splicing pathway (Park et al., Nucleic Acids Res., 2001, 29, 3595-3602).
Gemin5 (also known as DKFZP586M1824 protein; gem-associated protein 5) was found by coimmunoprecipitation of proteins that associate with SMN in vivo. Like SMN, Gemin 5 is localized in the cytoplasm, nucleoplasm, and is highly enriched in the nuclear gems. It binds to SMN by direct protein-protein interaction in vitro and interacts with several of the snRNP core Sm proteins. The Gemin5 protein is predicted to contain up to 13 WD repeats in its amino-terminal half, and a coiled-coil near its carboxyl terminus. Because both WD repeats and coiled-coil motifs are protein-protein interaction domains, Gemin5 may serve as a structural platform for protein assembly (Gubitz et al., J. Biol. Chem., 2002, 277, 5631-5636).
Extracts from stable cell lines expressing epitope-tagged SMN or Gemin2 proteins were analyzed by immunoprecipitation with an antibody recognizing the epitope. Both tagged-proteins were isolated with the SMN complex which was found to contain additional previously unidentified proteins including Gemin6 (also known as GEM-associated protein 6 or hypothetical protein FLJ23459). Database searches revealed that Gemin6 is not significantly homologous to other proteins and contains no common motifs which may be indicative of function. Like the other members of the SMN complex, Gemin6 was shown to interact with several Sm proteins (Baccon et al., J. Biol. Chem., 2002, 277, 31957-31962). The Gemin6 localization is similar to the other components of the SMN complex, but direct interaction of Gemin6 with SMN, Gemin2, Gemin3, Gemin4 or Gemin5 was not detectable (Pellizzoni et al., J. Biol. Chem., 2002, 277, 7540-7545).
However, Gemin7 (also known as hypothetical protein FLJ13956), also identified using the epitope-tagged system to purify SMN complexes, was shown to bind directly to Gemin6 and SMN in vitro, therefore it likely mediates the association of Gemin6 with the SMN complex. Like Gemin6, Gemin7 does not contain any known motifs that may suggest possible functions, but it does interact with a subset of the Sm proteins. Like the other complex members, Gemin7 colocalizes with SMN in gems (Baccon et al., J. Biol. Chem., 2002, 277, 31957-31962).
Beyond interactions with Sm proteins and snRNAs, the SMN complex interacts directly with several protein targets that are components of RNPs which function in various aspects of RNA metabolism. Among these substrates are the Sm-like (Lsm) proteins of the snRNPs, also essential components of the splicing machinery (Friesen and Dreyfuss, J. Biol. Chem., 2000, 275, 26370-26375). Fibrillarin and GAR1, components of small nucleolar RNPs (snoRNPs), also interact with SMN. Fibrillarin is a marker for Box C/D snoRNPs, the class that is necessary for cleavage and site-specific methylation of rRNA. Box H/ACA snoRNPs contain GAR1 and guide the pseudouridylation of rRNA (Pellizzoni et al., Curr. Biol., 2001, 11, 1079-1088). Thus, the SMN complex also appears to be involved in snoRNP biogenesis (Jones et al., J. Biol. Chem., 2001, 276, 38645-38651; Pellizzoni et al., Curr. Biol., 2001, 11, 1079-1088). Additional SMN complex substrates are hnRNP U and Q, RNA helicase A, and Epstein-Barr virus nuclear antigen 2 (EBNA2), coilin, and nucleolin (Barth et al., J. Virol., 2003, 77, 5008-5013; Gubitz et al., Exp. Cell. Res., 2004, 296, 51-56; Liu and Dreyfuss, Embo J., 1996, 15, 3555-3565; Mourelatos et al., Genes & Development, 2002, 16, 720-728; Yong et al., Trends Cell Biol., 2004, 14, 226-232). Because most SMN complex substrates are components of various RNP complexes involved in RNA processing, the SMN complex may take part in many aspects of cellular RNA metabolism (Gubitz et al., Exp. Cell. Res., 2004, 296, 51-56).
Disclosed in U.S. Pat. No. 6,646,113 is an antisense isolated nucleic acid complementary to the nucleic acid encoding a human Survival of Motor Neuron-Interacting Protein 1, wherein said nucleic acid encodes a protein that differs from the amino acid sequence disclosed by a mutation that inhibits binding of the Survival of Motor Neuron protein, and further wherein said mutation is selected from the group consisting of a deletion of the carboxyl terminal 89 amino acids relative to the amino acid sequence disclosed therein and a deletion of the carboxyl terminal 162 amino acids relative to the amino acid sequence disclosed therein. Also disclosed are antisense oligomers of between about 10 to about 30, and more preferably about 15 nucleotides (Dreyfuss et al., 2003).
U.S. Pre-Grant publication 20030228617 discloses a kit comprising a plurality of oligonucleotide primers and instructions for employing the plurality of oligonucleotide primers to determine the expression level of at least one of the genes represented in a group of sequences, said group including a nucleic acid sequence of a partial cDNA and a full-length cDNA corresponding to the human survival of motor neuron protein interacting protein 1 (SIP1) gene (Aune and Olsen, 2003).
The role of Gemin2, Gemin4, Gemin5, Gemin6, and Gemin7 in RNA metabolism makes these attractive targets for therapeutic and investigative strategies aimed at antisense technology. Consequently, there remains a need for agents capable of effectively modulating Gemin2, Gemin4, Gemin5, Gemin6, and Gemin7 function.
Antisense technology is an effective means for reducing the expression of one or more specific gene products and can therefore prove to be uniquely useful in a number of therapeutic, diagnostic, and research applications.
Consequently, there remains a long-felt need for agents that specifically regulate gene expression via antisense mechanisms. Disclosed herein are antisense compounds useful for modulating gene expression pathways, including those relying on mechanisms of action such as RNaseH, RNAi and dsRNA enzymes, as well as other antisense mechanisms based on target degradation or target occupancy. One having skill in the art, once armed with this disclosure will be able, without undue experimentation, to identify, prepare and exploit antisense compounds for these uses.